cortex 48 (2012) 46e57 Available online at www.sciencedirect.com Journal homepage: www.elsevier.com/locate/cortex Special issue: Review The prefrontal cortex: Comparative architectonic organization in the human and the macaque monkey brains Michael Petrides a,*, Francesco Tomaiuolo b,c, Edward H. Yeterian d,e,f and Deepak N. Pandya e,f,g a Montreal Neurological Institute, Department of Neurology and Neurosurgery, and Department of Psychology, McGill University, Montreal, QC, Canada b Auxilium Vitae Volterra, Volterra, Pisa, Italy c Department of Internal Medicine and Public Health, University of L’Aquila, Italy d Department of Psychology, Colby College, Waterville, ME, USA e Edith Nourse Rogers Memorial VA Medical Center, Bedford, MA, USA f Boston University School of Medicine, Boston, MA, USA g Beth Israel Deaconess Medical Center, Boston, MA, USA article info abstract Article history: Detailed cytoarchitectonic studies of the human cerebral cortex appeared during the first Received 23 February 2011 quarter of the 20th century. The incorporation of the cytoarchitectonic map by Brodmann Reviewed 18 March 2011 (1909) in the Talairach proportional stereotaxic space (Talairach and Tournoux, 1988) has Revised 7 April 2011 established the Brodmann numerical nomenclature as the basis for describing the cortical Accepted 19 July 2011 location of structural and functional findings obtained with modern neuroimaging. In Published online 29 July 2011 experimental anatomical and physiological investigations of the macaque monkey per- formed during the last 50 years, the numerical architectonic nomenclature used to describe Keywords: findings in the prefrontal cortex has been largely based on the map by Walker (1940). Prefrontal cortex Unfortunately, the map by Walker was not based on a comparative investigation of the Cytoarchitecture cytoarchitecture of the human and macaque monkey prefrontal cortex and, as a result, the Fiber pathways nomenclature and the criteria for demarcating areas in the two primate species are not always consistent. These discrepancies are a major obstacle in the ability to compare experimental findings from nonhuman primates with results obtained in functional and structural neuroimaging of the human brain. The present article outlines these discrep- ancies in the classical maps and describes comparative investigations of the cytoarchi- tecture of the prefrontal cortex of the macaque monkey and human (Petrides and Pandya, 1994, 1999, 2002a) in order to resolve these discrepancies and enable easy translation of experimental research in the monkey to findings in the human brain obtained with modern neuroimaging. ª 2011 Elsevier Srl. All rights reserved. * Corresponding author. Montreal Neurological Institute, Department of Neurology and Neurosurgery, 3801 University Street, Montreal, Quebec, Canada H3A 2B4. E-mail address: [email protected] (M. Petrides). 0010-9452/$ e see front matter ª 2011 Elsevier Srl. All rights reserved. doi:10.1016/j.cortex.2011.07.002 cortex 48 (2012) 46e57 47 The cerebral cortex can be divided into several distinct which they were located. The stereotaxic map of Talairach cytoarchitectonic areas on the basis of differences in cell size (Talairach and Tournoux, 1988) was adopted by the functional and type and in the arrangement of the neurons in the various neuroimaging community to provide a standard proportional cortical layers, such as differences in cell density, the pres- stereotaxic space within which all brains could be trans- ence or absence of certain layers, and the relative thickness of formed in order to correct, partially, for differences in size and the layers. The first complete cytoarchitectonic maps to be shape and permit reporting of the structural or functional published were those of Campbell (1905), who divided the neuroimaging findings in a common stereotaxic space (see human cerebral cortex into a few general regions, and the Brett et al., 2002). Gradually, the Talairach stereotaxic space, map of the monkey (Cercopithecus) cerebral cortex published which was based on one brain, evolved into the Montreal by Brodmann (1905). A little later, Brodmann (1908, 1909, 1914) Neurological Institute (MNI) standard proportional stereotaxic published his famous map of the human cerebral cortex. In space (MNI space) that was based on several brains (Collins Brodmann’s maps, several cortical areas were identified and et al., 1994). The Talairach space and its modern develop- labeled with distinct numbers (Figs. 1A and 2A). In 1925, ment, the MNI space, constitute now the common stereotaxic Economo and Koskinas published a major atlas of the human framework within which specific activity changes in func- cerebral cortex in which the different architectonic areas were tional neuroimaging studies and/or morphological changes in labeled with letters (Fig. 1B) and provided a detailed descrip- the brain as a result of training or disease are described. The tion of the different areas and excellent photomicrographs. In use of the Brodmann (1909) cytoarchitectonic numbers to the 1950s, the maps of Bailey and Bonin (1951) and Sarkissov describe the different areas of the cerebral cortex in the et al. (1955) appeared, the latter map being a modified Talairach and Tournoux (1988) atlas also meant the wide version of the Brodmann map based on the examination of adoption of the Brodmann cortical scheme by the functional several brains. Various maps focused on the cytoarchitecture neuroimaging community in the description of functional and of the human frontal lobe, such as the map by Sanides (1962), morphological changes in the human cerebral cortex. the map of the orbital frontal region by Beck (1949), the The identification of the cytoarchitectonic area within dorsolateral frontal areas 9 and 46 by Rajkowska and which the functional activity occurred is a complex issue that Goldman-Rakic (1995), Broca’s region by Amunts et al. (1999), requires careful consideration of many factors. Probability and areas 10 and 13 by Semendeferi et al. (1998, 2001). Apart maps which provide quantitative information about the from the cytoarchitectonic studies mentioned above, some location of particular structures in the Talairach or MNI space investigators described the architecture of the cerebral cortex have been published to aid the investigator in making deci- based on myelin (Vogt, 1910; Vogt and Vogt, 1919) or pigment sions about the locus of the activation. These probability maps architecture (Braak, 1979). have been of particular cytoarchitectonic areas (e.g., Amunts Architectonic studies of the human cerebral cortex were of et al., 1999) or morphological structures, such as the pars relatively limited interest until the emergence of modern opercularis (e.g., Tomaiuolo et al., 1999) or the orbitofrontal functional neuroimaging in the 1980s. The demonstration sulci (e.g., Chiavaras et al., 2001). In the latter case, the with positron emission tomography (PET), initially, and a little assumption has been that architecture maintains a more-or- later with functional magnetic resonance imaging (fMRI) that less stable relation to certain morphological entities, which focal changes in cortical activity could be detected in relation is known to be the case for some structures. For instance, the to various aspects of motor and cognitive performance primary visual area (the striate cortex) is known to lie within required a stereotaxic map to describe the location of these the banks of the calcarine sulcus, the primary motor cortex changes and to identify the cytoarchitectonic areas within (area 4) always lies in the anterior bank of the central sulcus Fig. 1 e The frontal cortex of the human brain as parcellated in the cytoarchitectonic maps of (A) Brodmann (1909) and (B) Economo and Koskinas (1925). 48 cortex 48 (2012) 46e57 Fig. 2 e Cytoarchitectonic map of the monkey cerebral cortex by (A) Brodmann (1905) and the prefrontal cortex by (B) Walker (1940). and the primary somatosensory cortical area 3 always lies in since the overall activity changes demonstrated by neuro- the posterior bank of the central sulcus. When such relations imaging within a region do not provide any information about are known, a reasonable inference can be made of the archi- the actual neuronal computations that underlie particular tectonic location of certain functional changes. Nowadays, functional contributions, examination of the physiological and cortical activation foci can be detected even in individual pharmacological properties of neurons recorded in these areas subjects and, therefore, the functional activations can be in alert behaving animals are necessary. related easily to particular sulci/gyri in an individual subject Since the 1950s, the development of experimental (e.g., Amiez et al., 2006). anatomical methods to study the connections between The frontal cortex, which is the focus of the present article, different brain areas, such as the silver degeneration (Fink and comprises several architectonic areas in both the human and Heimer, 1967; Nauta, 1957), initially, and later the anterograde the monkey brain (e.g., Barbas and Pandya, 1989; Brodmann, tracer methods (e.g., autoradiography, Cowan et al., 1972), 1905, 1908, 1909, 1914; Economo and Koskinas, 1925; Petrides made it possible to identify axonal fiber tracts in experimental and Pandya, 1999, 2002a; Sarkissov et al., 1955; Walker, 1940). animals. In such material, one can identify the precise course Experimental anatomical studies in the monkey have shown